Taxa Biodiversity Inventory

22 Pages·2015·4.25 MB·English

Organization, usefulness and limitations of an ATBI
(All Taxa Biodiversity Inventory): the inventory of terrestrial
invertebrates in the Mercantour National Park
Louis DEHARVENG
Anne BEDOS
Christophe DAUGERON
Claire VILLEMANT
Institut de Systématique, Évolution, Biodiversité,
UMR 7205 CNRS, MNHN, UPMC, EPHE,
Muséum national d'Histoire naturelle, Sorbonne Universités,
case postale 50, 57 rue Cuvier, 75005 Paris (France)
[email protected]
[email protected]
[email protected]
[email protected]
Mark L. I. JUDSON
Institut de Systématique, Évolution, Biodiversité,
UMR 7205 CNRS, MNHN, UPMC, EPHE,
Muséum national d'Histoire naturelle, Sorbonne Universités,
case postale 53, 57 rue Cuvier, 75005 Paris (France)
[email protected]
Published on 27 March 2015
urn:lsid:zoobank.org:pub:D10A950C-2F72-4592-9623-FC942E8AE32C
Deharveng L., Bedos A., Daugeron C., Villemant C. & Judson M. L. I. 2015. — Organization, usefulness and limitations
of an ATBI (All Taxa Biodiversity Inventory): the inventory of terrestrial invertebrates in the Mercantour National Park, in
Daugeron C., Deharveng L., Isaia M., Villemant C. & Judson M. (eds), Mercantour/Alpi Marittime All Taxa Biodiversity
Inventory. Zoosystema 37 (1): 9-30. http://dx.doi.org/10.5252/z2015n1a1
ABSTRACT
We present here the objectives, organization and preliminary results of the invertebrate inventory of
the Mercantour National Park (“Terrestrial Invertebrates Module” of the ATBI Mercantour/Alpi Marit-
time), carried out by 68 scientists and fi eld-collectors from 2009 to 2012. Th e conceptual framework
of an ATBI and the interactions between inventorying and frequently associated research aims are dis-
cussed. Th e sampling strategy adopted combined standardized multi-taxon sampling (continuous and
occasional) with individual taxon-centred sampling. Th e successive tasks of fi eld sampling, specimen
coding, sorting and identifi cation, molecular analysis and data management are briefl y described. A
coherent system of sample coding ensured the traceability of specimens. Th e project results in terms of
sampling eff ort, abundance of collected invertebrate fauna and biological diversity are presented globally,
by techniques, by sites and by higher taxonomic groups: 259,412 specimens were obtained from 761
sorted samples (out of 1243 samples collected), belonging to 1725 identifi ed species and subspecies, in
addition to which 37 species new to science have been recognized. Th ese data are progressively being
integrated into the CardObs database of the INPN and are partly accessible through the EDIT and
ZOOSYSTEMA • 2015 • 37 (1) © Publications scientifi ques du Muséum national d’Histoire naturelle, Paris. www.zoosystema.com 9
Deharveng L. et al.
GBIF websites. Molecular analyses (barcoding) have provided 2800 sequences (mostly for the COI
KEY WORDS gene in Gastropoda, Collembola and Hymenoptera). Analysis of these sequences indicates a signifi cant
Biodiversity,
amount of cryptic diversity in the material. From a macro-ecological perspective, endemicity in the park
fi eld-sampling,
labelling, decreases globally from south to north and species richness decreases at higher elevations, but the data
sampling strategy, are uneven and patterns vary according to group. Based on this four-year experience, it appears that a
sampling techniques,
southern Alps, smaller survey area with lower habitat diversity would be better suited to the aims of an ATBI when
specimen traceability. the available resources are limited, and would enhance the effi ciency of collective work in the fi eld.
RÉSUMÉ
Organisation, intérêt et limites d’un ATBI (All Taxa Biodiversity Inventory) : l’inventaire des invertébrés
terrestres dans le Parc national du Mercantour.
Nous présentons les objectifs, l’organisation et les résultats préliminaires de l’inventaire des inver-
tébrés du Parc National du Mercantour (« Module Invertébrés Terrestres », ATBI Mercantour/Alpi
Marittime), eff ectué par 68 scientifi ques et collecteurs de terrain de 2009 à 2012. Le cadre conceptuel
d’un ATBI, et les interactions entre un inventaire et les approches scientifi ques qui lui sont souvent
associées sont discutés. La stratégie d’échantillonnage adoptée associe des collectes standardisées
multi-taxon (en continu et ponctuelles) avec des collectes individuelles taxon-centrées. Les tâches
successives – échantillonnage, codage des spécimens, tri et identifi cations, analyses moléculaires,
gestion des données – sont brièvement décrites. Le codage assure la traçabilité des spécimens. Les
résultats du projet (échantillonnage, abondance des invertébrés collectés, diversité biologique) sont
présentés globalement, par techniques, par sites et par groupes taxonomiques: 259 412 spécimens
ont été obtenus dans les 761 échantillons analysés (sur les 1243 collectés) appartenant à 1725 espèces
et sous-espèces identifi ées, auxquelles s’ajoutent 37 espèces nouvelles pour la science. Ces données
sont en cours d’intégration dans la base CardObs (INPN, France) et en partie accessibles (sites web,
EDIT et GBIF). Les analyses moléculaires (barcode) ont fourni 2800 séquences (Gastéropodes,
Collemboles, Hyménoptères, essentiellement COI): leur analyse suggère une diversité cryptique
MOTS CLÉS importante au sein du matériel étudié. Du point de vue macroécologique, l’endémicité dans le parc
Biodiversité,
décroit globalement du sud au nord, et la richesse spécifi que des basses aux hautes altitudes, mais les
échantillonnage de terrain,
étiquetage, données sont hétérogènes et les patrons varient selon les groupes. Sur la base de cette expérience de
stratégie d’échantillonnage, quatre ans, il apparait qu’une zone d’étude plus réduite que l’ensemble du PNM avec une moindre
techniques d’échantillonnage,
diversité d’habitats serait mieux adaptée aux objectifs d’un ATBI compte tenu des ressources dispo-
Alpes du sud,
traçabilité de spécimens. nibles, et améliorerait l’effi cacité du travail collectif sur le terrain.
INTRODUCTION (Vasarhelyi 2007). In the Massane forest, a natural reserve
of 336 ha created in 1973, 6556 species have been recorded
Th e aims of biological inventories are to determine the level so far, accompanied by abundant data on their ecology and
and patterns of biodiversity in a given area, i.e. the number of biology (Garrigue 2014).
taxa, their distribution, the diversity of their biological traits Th e Mercantour/Alpi Marittime inventory project (“M-
and the diversity of their assemblages. Such inventories have AM” project) in the southwestern Alps was designed in a
experienced a recent renewal (Mauz 2011) and are currently similar way, but at a much larger geographical scale, since the
being conducted around the world by academic scientists, territory concerned is, for example, 700 times larger than the
NGOs and amateurs, at various taxonomic, geographic and Massane forest. Coordinated by the Museum für Naturkunde
temporal scales. However, massive inventories, involving a of Berlin and the State Museum of Natural History of Stutt-
wide range of taxa, remain exceptional (Baldi 1999). Th ey gart, as a workpackage of the European project EDIT (http://
range from large expeditions that mobilize numerous special- www.atbi.eu/mercantour-marittime), it was the fi rst “All Taxa
ists over several months, such as the Santo 2006 expedition Biodiversity Inventory” (ATBI) – a term coined by Janzen &
(Bouchet et al. 2007), to long-term studies encompassing Hallwachs (1994) – of this programme to be implemented
a large diversity of taxa in a well delimited region, such as in Europe, inspired in part by the large scale inventory of the
the ALAS project in Costa-Rica, which started in 1991 and Great Smoky Mountains (Sharkey 2001; Nichols & Langdon
is still running (http://viceroy.eeb.uconn.edu/alas/ALAS. 2007; Bernard & Felderhof 2007). Th e aim of the project
html). In Europe, possibly the most successful inventories was to inventory as completely as possible the biodiversity
of this type are those of the Bükk National Park in Hungary of the Mercantour and Alpi Marittime parks, initially at just
and the Massane forest in the Pyrénées-Orientales (France). a few reference spots, but later over throughout of the parks
For the Bükk National Park, 9436 animal species are listed, (Mauz & Granjou 2013). Two smaller-scale ATBIs were also
based on a massive inventory conducted from 1981 to 1985 established within EDIT, in Slovakia and Germany. EDIT
10 ZOOSYSTEMA • 2015 • 37 (1)
Invertebrates inventory in the Mercantour ATBI
ended in February 2011, but further funding was obtained ATBI
Scientific Public
by the parks through a European trans-border programme, outputs outputs
ALCOTRA.
Sampling
Th e Mercantour and Alpi Marittime parks together extend
Taxonomy
between latitudes 43.90° to 44.40°N, and longitudes 6.60° to Sorting
7.52°E, covering almost 100 000 ha, i.e. a large part of one of Faunistics/ Open access
the richest hotspots of biodiversity in Europe, the southwestern biogeography Identification databases
Alps. With several peaks above 3000 m (Monte Argentera, at
Ecology Diffusion
3297 m, and Cime du Gelas, at 3143 m, being the highest), List (inventory)
an altitudinal range of more than 2500 metres, a contrasted Genetics
climate mixing Mediterranean and Alpine infl uences, and
geological terrains ranging from limestone to crystalline, the
Monitoring Collections
habitats of the southwestern Alps are extremely diverse. Because
this region was located at the southern limit of the glaciers
during the last glaciation, it has retained a large number of FIG.1. — Main goals and context of an ATBI (after Deharveng & Isaia 2013).
endemic plants (Medail & Quezel 1997; Pawlowski 1970)
and invertebrates (Sainte-Claire Deville 1928; Deharveng et
al. 2000). Beside the interest of targeting such a remarkable 4) explore cryptic biodiversity in target groups through
hotspot of biodiversity, the M-AM project represents a de- barcoding;
parture from traditional biodiversity inventories in its large 5) reconstruct patterns of biodiversity and endemism across
scale, in terms of taxonomic scope, size of studied area and the study area;
number of contributing taxonomists. Terrestrial invertebrates 6) document biodiversity changes along altitudinal and
are of particular interest in this context, being by far the most latitudinal gradients;
diversifi ed group of organisms (Chapman 2009) and the rich- 7) develop projects on the cryophilous fauna and on pol-
est in terms of endemics. Th e Terrestrial Invertebrate Module linisation (not considered here);
(“TIM”) of the project ran for four years (2009-2012), with 8) provide validated datasets to the Inventaire National
fi eldwork conducted from 2009 to 2011 in the Mercantour du Patrimoine Naturel (France) database for public release;
National Park (MNP) area. 9) enrich the collections of the Muséum national d’Histoire
Th is paper provides an overview of the operational or- naturelle (Paris, France).
ganization of the inventory of the terrestrial invertebrates Th e ATBI projects of EDIT were initially designed to pro-
(excluding cavernicolous species) conducted by the TIM in vide a basis for biodiversity monitoring activities over time
the Mercantour National Park and summarizes the scientifi c (hence the “M” in ATBI+M) from a baseline inventory, but
results obtained so far. this objective has yet to be implemented.
OBJECTIVES (FIG. 1) MATERIAL AND METHODS
Th e aims of an ATBI are to develop an inventory, as compre- ACRONYMS AND DEFINITIONS
hensive as possible, of the plants and animals living in a given ATBI All Taxa Biodiversity Inventory;
ATBI+M All Taxa Biodiversity Inventory + Monitoring, offi cial
territory, including data on the environment and location of
acronyme of the M-AM ATBI, workpackage of EDIT;
collection sites, and the dates of sampling. Th e complete-
CardObs an INPN database in which the Mercantour distribu-
ness of an inventory depends directly on the diversity of the tion data are stored;
chosen habitats and the methods used to sample them. In EDIT European Distributed Institute of Taxonomy;
practice, however, because ecological and genetic objectives INPN Inventaire national du Patrimoine naturel (French
National Inventory of Natural Heritage);
are usually added to such inventories, replicates are needed,
M-AM Mercantour-Alpi Marittime;
resulting in a lower diversity of sampled habitats and hence a
MNHN Muséum national d’Histoire naturelle, Paris;
lower number of species obtained for a given sampling eff ort. MNP Mercantour National Park;
However, integrating such peripheral objectives, which are not SSM Service de systématique moléculaire of the MNHN;
those of an ATBI, does allow investigation of the origin and TIM Terrestrial Invertebrate Module;
Sample a collecting event defi ned by its date, its sampling
environmental connections of the observed diversity patterns
technique(s), its location and its collectors;
(Deharveng & Isaia 2013).
Record presence of a given taxon in a single sample.
Th e aims of the TIM were to:
1) carry out a basic inventory of the terrestrial invertebrate WORK ORGANIZATION (FIG. 2)
fauna; Th e ATBI participants contributed either individually or
2) detect and describe taxa new to science; through modules to the inventory. Each module dealt with a
3) initiate a barcode library for invertebrates of certain particular component of biodiversity: terrestrial (non-subterra-
target groups; nean) invertebrates, aquatic (non-subterranean) invertebrates,
ZOOSYSTEMA • 2015 • 37 (1) 11
Deharveng L. et al.
subterranean terrestrial fauna, subterranean aquatic fauna, TABLE 1. — Terrestrial Invertebrates Module of the Mercantour/Alpi Marittime
wild bees, fungi, lichens, bryophytes and vascular plants. Th e ATBI: taxonomic coverage and taxonomists involved, Occasional contributors
in square brackets.
TIM, which was the largest module of the project, focused
on non-subterranean terrestrial invertebrates.
Taxon Specialist(s) in charge
Participants and contributors Annelida: Lumbricidae Sandrine Salmon (France)
Th e TIM was launched and coordinated by taxonomists Arachnida: Araneae Christophe Hervé (France),
Christine Rollard (France)
of the MNHN. Our fi rst task was to set up a network of
specialists for fi eld collections and for identifi cations, with a Arachnida: Opiliones Emmanuel Delfosse (France)
large taxonomic coverage and minimal overlap in expertise. Arachnida: Pseudoscorpiones Mark Judson (France)
A total of 68 researchers, students and PNM staff from 13 Collembola:
diff erent countries were associated with the project (Ta- Brachystomellidae Louis Deharveng (France)
Cyphoderidae Jantarit Sopark (Thailand)
ble 1). Of these, 25 participated in fi eld collecting and 62
Entomobryidae Galina Busmachiu (Moldova)
undertook identifi cations during the 4 years of the TIM Hypogastruridae Anne Bedos (France)
(2009-2012). As refl ected in this special issue of Zoosystema, Isotomidae Mikhail Potapov (Russia)
Neanuridae Louis Deharveng (France)
scientifi c work on the material continues and will do so
Neelidae Louis Deharveng (France)
for many years to come, with additional taxonomists and Odontellidae Louis Deharveng (France)
ecologists becoming involved. As can be seen from Table 6, Oncopoduridae Louis Deharveng (France)
Onychiuridae Xin Sun (China)
several major groups still have very few species identifi ed,
Symphypleona Louis Deharveng (France)
or even none at all. Because specialists are not currently Tomoceridae Daoyuan Yu (China)
available for a number of groups, it is clear that a large Tullbergiidae Wanda Weiner (Poland)
proportion of the biodiversity will remain undocumented Diplopoda:
for some time to come. Blaniulidae Jean-Jacques Geoffroy (France)
Callipodidae Jean-Jacques Geoffroy (France)
Craspedosomatidae Jean-Jacques Geoffroy (France)
Workfl ow summary Glomeridae Jean-Jacques Geoffroy (France)
Given the aims of the inventory, the massive collecting tech- Julidae Jean-Jacques Geoffroy (France)
Penicillata Duy Nguyen Monique (France)
niques used, and the large number of specialists involved, we
Polydesmidae Jean-Jacques Geoffroy (France)
had to organize the workfl ow from fi eld trips to deposition in
Gastropoda Olivier Gargominy (France)
collections in a fairly tightly constrained manner, comparable
Insecta: Coleoptera:
to the MarBOL workfl ow in MNHN described by Puillandre
Anobiidae Thomas Théry (France)
et al. (2012). Th e main tasks performed are briefl y summarized Buprestidae Jean Raingeard (France)
below (the diff erent aspects are described in more detail in Cantharidae Robert Constantin (France)
Carabidae Jacques Coulon (France),
the relevant sections).
Arnaud Faille (France),
Sampling and initial sorting phase. Sampling activities within Jean-Michel Lemaire (France)
the module were organized into three components operating Cerambycidae Jean Raingeard (France)
Cerylonidae Hervé Bouyon (France)
complementary sampling methods. Th e fi rst component, co-
Chrysomelidae Serge Doguet (France)
ordinated by Christophe Daugeron, mostly sampled fl ying Coccinellidae Jean-Pierre Coutanceau (France)
insects, using Malaise traps, interception traps, light traps and Curculionidae Laurent Schott (France)
Dascillidae Thomas Théry (France)
hand-netting. Th e second component, coordinated by Louis
Dermestidae Hervé Bouyon (France)
Deharveng, mainly sampled soil invertebrates, using pitfall Elateridae Hubert Piguet (France)
traps, extraction with Berlese funnels and hand collecting. Eucnemidae Lucien Leseigneur (France)
Histeridae Thomas Théry (France)
Th e third component, coordinated by Olivier Gargominy, was
Hydraenidae Pierre Queney (France)
devoted to snails. Several participants also worked indepen- Hydrophilidae Pierre Queney (France)
dently in the fi eld and contributed to the collective approach Lathridiidae Bernard Moncoutier (France)
of the module by surveying reference sites and identifying the Lycidae Roland Allemand (France)
Lymexylidae Thomas Théry (France)
material obtained.
Melandryidae Hervé Bouyon (France)
Th e fi rst step after sampling consisted of sorting specimens Meloidae Thomas Théry (France)
into the diff erent classes and orders of invertebrates. Th is time- Melyridae Robert Constantin (France)
Mordellidae Pascal Leblanc (France)
consuming task was supported during three years by specifi c
Nitidulidae Bernard Moncoutier (France)
conventions with the MNP, the Collections Service of the Oedemeridae Thomas Théry (France)
MNHN and the Systematics and Evolution Department of Ptininae Roland Allemand (France)
Pyrochroidae Thomas Théry (France)
the MNHN. It was carried out by dedicated temporary staff
Scarabaeoidea Olivier Montreuil (France)
with a good knowledge of general insect taxonomy, under Scolytidae Thierry Noblecourt (France),
the supervision of group coordinators. [Thomas Théry (France)]
Scraptiidae Pascal Leblanc (France)
Second sorting phase and identifi cation. While each sampling
Silphidae Thomas Théry (France)
method is most effi cient for a certain type of invertebrates, it Sphindidae Thomas Théry (France)
can also yield signifi cant proportions of non-target species. Staphylinidae Marc Tronquet (France)
12 ZOOSYSTEMA • 2015 • 37 (1)
Invertebrates inventory in the Mercantour ATBI
TABLE 1. — Continuation.
Taxon Specialist(s) in charge Taxon Specialist(s) in charge
Insecta: Coleoptera: Insecta: Hymenoptera:
Tenebrionidae Hervé Bouyon (France), Sphecidae Pierre Tripotin (France)
[Thomas Théry (France)] Symphyte David R. Smith (USA)
Tetratomidae Hervé Bouyon (France) Vespidae Quentin Rome (France),
Throscidae Lucien Leseigneur (France) [Phil Lester (New Zealand)]
Trogositidae Thomas Théry (France)
Insecta: Lepidoptera:
Insecta: Dermaptera: Forfi culidae Thomas Théry (France) Amphisbatidae Patrice Leraut (France)
Arctiidae Patrice Leraut (France)
Insecta: Diptera:
Coleophoridae Patrice Leraut (France)
Acroceridae Patrice Leraut (France)
Crambidae Patrice Leraut (France)
Bombyliidae Patrice Leraut (France)
Epermeniidae Patrice Leraut (France)
Conopidae Axel Ssymank (Germany),
Gelechiidae Patrice Leraut (France)
[Patrice Leraut (France)]
Geometridae Antoine Lévêque (France),
Dolichopodidae Marc Pollet (Belgium)
[Patrice Leraut (France)]
Drosophilidae Jean David (France)
Lycaenidae Patrice Leraut (France)
Empidinae Christophe Daugeron (France),
Momphidae Patrice Leraut (France)
[Vincent Lefebvre (France)]
Noctuidae Jérôme Barbut (France),
Pipunculidae Christian Kehlmaier (Germany)
[Patrice Leraut (France)]
Stratiomyidae Axel Ssymank (Germany)
Nymphalidae Patrice Leraut (France)
Syrphidae Axel Ssymank (Germany)
Oecophoridae Patrice Leraut (France)
Tachinidae Patrice Leraut (France)
Plutellidae Patrice Leraut (France)
Insecta: Hemiptera: Prodoxidae Patrice Leraut (France)
Auchenorrhyncha Adeline Soulier-Perkins (France) Pterophoridae Patrice Leraut (France)
Cicadellidae Gernot Kunz (Austria) Pyralidae Patrice Leraut (France)
Coccoidea Danièle Matile (France) Scythridae Patrice Leraut (France)
Psyllidae David Ouvrard (France) Sesiidae Patrice Leraut (France)
Tortricidae Patrice Leraut (France)
Insecta: Heteroptera: Miridae Armand Matocq (France)
Yponomeutidae Patrice Leraut (France)
Insecta: Hymenoptera:
Insecta: Mecoptera Pierre Tillier (France)
Apidae Franck Muller (France)
Cynipoidea Juli Pujade-Villars (Spain) Insecta: Neuroptera Pierre Tillier (France)
Dryinidae Massimo Olmi (Italy)
Insecta: Rhaphidioptera Pierre Tillier (France)
Formicidae Anthony Turpaud (France),
Quentin Rome (France) Insecta: Strepsiptera Hans Pohl (Germany)
Ichneumonidae Claire Villemant (France),
Insecta: Trichoptera:
Takuma Yoshida (Japan)
Limnephilidae Patrice Leraut (France)
Pompilidae Frédéric Durand (France)
Philopotamidae Patrice Leraut (France)
Th us, Malaise and interception traps provided, in addition Data management. In addition to supervising the second
to a huge number of fl ying insects, many wingless inverte- sorting step for their group, each coordinator had to: 1) or-
brates (such as arachnids, springtails, insect larvae and ants) ganize the dispatching of specimens to relevant specialists
that belong to taxonomic groups that are largely represented for identifi cation; 2) recover identifi cations and specimens;
in the soil. Th e second step in the sorting tasks was therefore 3) manage and clean datasets in the project database; 4) ex-
organized diff erently from the fi rst, with coordinators being tract and send validated subsets of data to the INPN data-
appointed for each order or class. Under their supervision, base; 5) send specimens of the target groups for barcoding
and irrespective of the sampling method, specimens were and 6) deposit specimens in the collections. Th e amount of
sorted to operational taxonomic levels for identifi cation, work required for the coordination of these tasks across the
depending on the availability of specialists for the groups diff erent supervisors (Fig. 2), the management and storing
concerned. For example, sorting was carried out to family of a rapidly increasing amount of material and data of vari-
or superfamily for beetles, where specialists usually focus on ous origins, and dealing with the administrative constraints
only one or a few families, and to ordinal level for arachnids linked to the conventions that supported our work was grossly
(except mites) and myriapods, where specialists usually deal underestimated in the initial stages. Th is led us to rapidly set
with one or more orders. up basic management rules with a dedicated supervising staff ,
Barcoding. DNA barcoding was used to complement tradi- including standardization of protocols for data communica-
tional taxonomic methods and investigate the possible pres- tion and specimen storage.
ence of cryptic species or diff erentiation between populations.
After identifi cation and databasing of associated information, SAMPLING STRATEGY
selected specimens were transmitted to the SSM for extrac- Following experience from large biological expeditions, such
tion and sequencing (see section “Processing specimens for as Santo 2006 (Bedos, Prié & Deharveng 2011; Deharveng &
barcoding”). Sémah 2011), and from various campaigns of terrestrial
ZOOSYSTEMA • 2015 • 37 (1) 13
Deharveng L. et al.
SSaammpplliinngg ddeessiiggnn
OOnnee--sshhoott ccoolllleeccttiinngg
FFiieelldd
ssaammpplliinngg
FFiieelldd ttrriippss aanndd ssaammpplliinnnggg CCoonnttiinnuuoouuss ccoolllleeccttiinngg
mmaannaaggeemmeenntt
SSaammppllee ccooddiinngg
LLaabb wwoorrkk
SSoorrttiinngg mmaannaaggeemmeenntt
FFaauunnaall eexxttrraaccttiioonn ((ssooiill))
SSttoorraaggee mmaannaaggeemmeenntt
SSaammppllee rreeccoonnddiittiioonniinngg
gg
nn LLooaann ffoollllooww--uupp
sisi
aa SSoorrttiinngg pphhaassee 11 bbyy hhiigghheerr ttaaxxaa
abab IIddeennttiiffiiccaattiioonnss ffoollllooww--uupp
atat
DD DDaattaa qquuaalliittyy ccoonnttrrooll SSoorrttiinngg pphh.. 22 aatt ooppeerraattiioonnaall rreessoolluuttiioonn
BBaarrccooddiinngg aanndd ssppeecciimmeenn
DDiissppaattcchhiinngg ffoorr iiddeennttiiffiiccaattiioonnss
mmaannaaggeemmeenntt
DDaattaa eexxppoorrttaattiioonn PPrreeppaarraattiioonn//llaabbeelllliinngg//ccoonnddiittiioonniinngg
SScciieennttiiffiicc eexxppllooiittaattiioonn ooff tthhee ddaattaa:: ttaaxxoonnoommyy,, iinnvveennttoorriieess,, mmaaccrrooeeccoollooggyy ........
IINNPPNN DDaattaabbaassee
((CCaarrDDoobbss)) BBaarrccooddiinngg MMNNHHNN ccoolllleeccttiioonnss
FIG. 2. — Organization of tasks in the module “Terrestrial Invertebrates” of the Mercantour-Alpi Marittime ATBI (blue frame) and connections with the different
MNHN services (orange frames).
biodiversity inventory involving multiple participants, we procedure, after which they were sorted for dispatching to
combined two strategies for sampling terrestrial invertebrates: relevant specialists (see above). In spite of producing mas-
independent sampling to catch the highest possible represen- sive collections, the integrated sampling techniques did not
tation of biodiversity, and standardized sampling to produce include attractive devices (light traps were only used during
comparable datasets. independent collecting).
Independent sampling was carried out by each specialist,
who will know better than anyone else which habitats and SAMPLING TECHNIQUES
methods will produce the highest diversity of species for A number of techniques are available for sampling terrestrial
his or her group. However, the drawback of this approach invertebrates (Achterberg et al. 2010; Häuser & Riede 2015).
is that it does not allow standardized comparisons between Each has its fi eld of application in terms of the spectrum of
the diff erent groups. Independent sampling is very effi cient taxa sampled, with frequent overlap and diff erent productive-
for the taxa of interest to the specialist, who tends however ness. Several techniques known for their high effi ciency and
to ignore other taxa, or at best collects and distributes only simplicity were used, in order to cover the maximum number
some of them to other specialists. During the project, indi- of species and habitats for minimal sampling eff ort and over-
vidual collecting was very productive and clearly showed that lap in the results. Th ey can be grouped into two categories
taxon-driven sampling is essential for completing a faunistic – continuous sampling and one-shot sampling – according to
inventory and gathering biological data, in spite of the huge the way in which they were implemented during the project.
amount of information potentially useful for the inventory
that is passed over in the process. Continuous sampling techniques (Fig. 3)
Integrated sampling used several complementary techniques Contrary to non-insect arthropods and soil insect larvae,
applied to selected sites, in order to obtain a comprehensive most adult insects can fl y and have short periods of activity
representation of the biodiversity at each site, along with that vary according to the species, from early spring to late
quantitative data for comparing biodiversity patterns. autumn. For this reason, continuous sampling was necessary
Th e sampling techniques were standardized, and the to obtain a correct overview of their diversity, given that our
sampling design identical at the diff erent sites. All speci- knowledge of the phenology of the species of the southwest-
mens collected were labelled and stored according to a fi xed ern Alps species is very limited. Th ree sampling methods
14 ZOOSYSTEMA • 2015 • 37 (1)
Invertebrates inventory in the Mercantour ATBI
were used to ensure such continuous sampling, with traps the debris is re-processed using a Berlese funnel. Th is tech-
changed every 15 days: nique was largely used during the project for small snails and
1) Malaise traps. Th ese are effi cient for fl ying insects (except low-density arthropods, such as pseudoscorpions.
Lepidoptera), for epigean Collembola, spiders and harvestmen. 4) Vegetation beating and sweep-netting. Beating plants,
Th e method is well known, robust and extremely productive. branches or shrubs on a white sheet or tray and sweeping
It mobilized most of the sorting work of the module. grasses with a net are the best methods for collecting species
2) Interception traps. Th ese are effi cient for fast-fl ying insects, living on the vegetation, which includes many winged insects,
especially beetles, which usually fall after hitting an obstacle. as well as arachnids and springtails. Specimens are collected
Th ey are also very productive, but less so than Malaise traps. directly from the tray, or the debris obtained is re-processed
Th e results they gave during the project were uneven. on Berlese funnel. Th is technique was occasionally used dur-
3) Pitfall traps containing 90% ethanol. Th ese are effi cient ing the project, particularly for spiders.
for surface soil fauna, particularly spiders, terrestrial isopods, 5) Light traps. Light trapping is the most productive col-
springtails, ants and other insects foraging at ground level, lecting method for moths and certain other groups of fl ying
such as ground-beetles. A single pitfall of 1 cm aperture insects. It was operated by individual taxonomists and only
typically provided 10-20 species in 15 days. However, the moths were collected. It was the only “attractive method”
interval between trap changes (15 days), imposed by the employed during the project. Since it was not used inten-
limited availability of human resources in the fi eld during sively, it is unlikely to have had a signifi cant impact on the
the sampling season, proved to be too long (see section con- local diversity of moths.
tinuous sampling). 6) Lumbricidae collecting. Earthworms were collected by
hand-picking after digging with a spade a surface of 0.1 m2
One-shot sampling techniques (Figs 4; 5) delimited using a metal cylinder, since the chemicals usually
Such methods do not rely on a permanent device in the fi eld. used to expel earthworms from soil are not allowed in the MNP.
Th ey can be used periodically at a given site during favourable 7) Hand collecting. Th is consists of searching for animals
seasons, provided that suffi cient human resources are avail- “by sight” under stones, logs or the bark of living trees. Th e
able. Since this was not the case for our project, they were animals found in this way are collected using a pooter (aspira-
applied for only one period per site during the three years of tor), a fi ne brush dipped in alcohol or, if they are large enough,
the TIM sampling. by simply picking them up with fi ngers. Such methods are
1) Extraction of soil cores (leaf litter and underlying soil) useful for the rapid collection of groups such as snails, mil-
using simplifi ed Berlese funnels (Fig. 5). Th is is the most lipedes, large arachnids, isopods, Diplura, large Collembola
practical and effi cient method for extracting soil arthro- and various insects, and it can provide species not obtained
pods. Because almost all mobile stages of soil arthropods with other methods (e.g. corticolous pseudoscorpions). Visual
living in a given soil sample are collected, this is the method searches are also useful for selecting substrates for subsequent
best suited to standardized approaches. Th e device used by extraction on Berlese funnels (e.g. rotting wood containing
the TIM consisted of a plastic funnel of 30 cm diameter, a Collembola). Since most of the nationally notable species
plastic mesh (2 mm mesh size, with a few larger holes added of snails present in the Mercantour Park are of large or very
peripherally) placed slightly below the rim of the funnel, and large size, hand collecting was largely favoured for this group,
a reception vial half-fi lled with 95% ethanol placed under its particularly in the context of cooperation with the park guards
stem. Th e freshly sampled soil core is placed on the grid and assisting with the collecting.
it dries progressively from its top, causing motile arthropods
to pass through the mesh and fall into the vial below. Th is Missing sampling techniques
technique was widely used during the project. Th e extraction Th e selection of sampling techniques was based on the fol-
was performed in a dry room, but no light was placed above lowing criteria: proven productivity, complementarity, ease
the sample to avoid killing sensitive specimens before they of implementation, minimal environmental impact, fi nancial
could fall into the vial. A set of 50 Berlese funnels was run viability and the availability of specialists to operate them.
simultaneously. Several methods were not adopted because they have very
2) Soil fl otation. Th is is used instead of Berlese extraction restricted applications (concerning only a minor component
when soil is too humid or contains a high proportion of fi ne of the fauna), examples being chemical attractants like phe-
mineral grains. Th e substrate is washed in a plastic basin and romonal traps. Others were excluded because they were not
fl oating arthropod specimens (mostly Collembola) are picked allowed in the protected area under study, an obvious exam-
up with a small brush. Alternatively, all fl oating material can ple being fogging with insecticides. One method that could
be skimmed off with a fi ne net and re-processed with a Berlese have provided access to important additional components of
funnel. Non-fl oating arthropods (e.g. terrestrial isopods) are arthropod biodiversity is acoustic surveying, but this was not
not obtained using this method. Th is technique was occasion- employed by the TIM due to the unavailability of a dedicated
ally used during the project. specialist during the project. Metabarcoding is also a promis-
3) Litter sifting. Sifting litter on a tray or using a Winkler ing tool for biodiversity inventorying, but this technique is
apparatus is an effi cient technique for concentrating the fauna. not yet fully operational for routine sampling and it currently
Specimens are collected directly from the sifted debris, or else lacks adequate DNA reference libraries.
ZOOSYSTEMA • 2015 • 37 (1) 15
Deharveng L. et al.
A
B C
D
FIG. 3. — Continuous collecting devices: A, Malaise trap (Larche, 2000 m); B, Malaise trap (Caïros, 2000 m); C, interception trap (Caïros, 2000 m); D, pitfall trap
(Caïros, 1400 m). Photographs: A, Emmanuel Delfosse; B, C, Quentin Rome; D, Louis Deharveng.
16 ZOOSYSTEMA • 2015 • 37 (1)
Invertebrates inventory in the Mercantour ATBI
A B
C D
E
G
F
FIG. 4. — One-shot sampling methods: A, earthworm collecting from core sample; B, collecting soil sample for Berlese extraction; C, sifting and hand collecting
with a fi ne brush from residues; D, litter sifting; E, soil fl otation; F, G, hand-netting. Photographs: A, Louis Deharveng; B, C, E, G, Donghui Wu; D, Olivier Gar-
gominy; F, Christophe Daugeron.
SAMPLING SITES (FIG. 6) diversity, and ecological objectives that required duplicates and
Sampling sites for the independent sampling strategy were standardization. Two major gradients shape biodiversity patterns
freely selected by the specialist concerned. As a result, they in the southwestern Alps: the latitudinal gradient of increasing
are irregularly scattered within the MNP. species richness from north to south, partly explained by last
For the integrated sampling, the rationale for site selection was a glaciation infl uence (Médail & Diadema 2009), and the altitu-
balance between pure inventory objectives for maximizing sample dinal gradient, typical of mountainous regions. Th e multi-taxon
ZOOSYSTEMA • 2015 • 37 (1) 17
Deharveng L. et al.
FIG. 5. — Extraction of litter sample using Berlese funnels. Photograph: Louis Deharveng.
approach adopted by the TIM represented an attempt to test the positioning of traps: they had to be not too visible from
and illustrate these gradients, which had not been documented public paths; the liquid preservative of the device should
previously for such a large array of zoological groups. Sampling be non-toxic for humans, not strongly coloured and, when
eff ort was equally distributed in four valleys, arranged along the employed in large amounts (interception traps), inaccessible
latitudinal gradient from north to south: the Larche valley in to cattle and children. Th e following traps were used at each
Ubaye, the Sestrière valley in Haute-Tinée, the Salèse valley near site and altitude:
Le Boréon, connected to the larger Vésubie valley, and the Caïros – two Malaise traps, placed 80-120 metres apart;
valley, connected to the larger La Roya valley (Fig. 6; Tables 2, – one or two interception traps;
3). In each valley, two sites were chosen, one at 1400-1500 m – fi ve pitfall traps randomly positioned in a ½ ha area.
and one at 2000 m. Sites at the same altitude were located in Every two weeks, staff of the MNP emptied and refi lled traps
similar vegetation types. It was initially planned to sample the with 90% ethanol (Malaise and pitfall traps) or with glycol
four valleys during the same season. Th is idea was scientifi cally (interception traps), from spring to autumn of each year. Th e
sound, but proved to be unrealistic given the fi nancial and hu- collection duration varied from year to year, depending on
man resources available. We therefore had to divide the fi eldwork the climatic conditions (mainly snowfall) at the beginning
over three years: 2009 (Boréon and Sestrière), 2010 (Caïros) and and end of the sampling period (Table 2). After topping-up
2011 (Larche). the levels of preserving fl uid when necessary, the samples were
sent within one week to the MNHN. A few samples were
SAMPLING DESIGN AT REFERENCE SITES not recovered in time because of bad weather conditions and
An integrated sampling design was applied to eight sites, others were destroyed by cows or humans, but overall 90% of
which were equipped with the same set of sampling devices the samples were retrieved. However, a signifi cant proportion
and were sampled following the same protocols. of pitfall traps contained damaged specimens because their
contents had decomposed (due to rain dilution, high tem-
Continuous sampling peratures or too long intervals between changing traps). Th e
Th e samples of this category obtained during the project are use of monopropylene glycol was considered, but it would
listed in Table 2A. Th e MNP placed moderate constraints on have posed other problems, since Collembola, the dominant
18 ZOOSYSTEMA • 2015 • 37 (1)